Lead AI Engineer

Onto Innovation is building AI systems for semiconductor metrology inspection and decision-making, with an emphasis on trust, provenance, and on-premises (including air-gapped) inference. This on-site role is hands-on and leadership-focused: you will shape the reference architecture, write production code, and guide teams as systems move from evaluation to real production lines measuring wafers around the clock.

What you’ll do

• Lead AI strategy and architecture for metrology: define what to build vs. buy, and keep the roadmap aligned to the customer problems that matter.
• Write and ship production code across prototypes, agent workflows, evaluation tooling, and the systems that run them.
• Design agent workflows for the full metrology lifecycle, including guided recipe and model setup, qualification runs, tool matching, and root-cause triage when measurements drift.
• Integrate with simulators, solvers, and tool data so agents can orchestrate steps while physics computes the result.
• Ground agents in fab knowledge using retrieval over manuals, recipes, service tickets, and historical measurement outcomes, with provenance on every answer.
• Build for trust with human-in-the-loop checkpoints, end-to-end confidence and uncertainty, complete audit trails, and refusal behavior when evidence is insufficient (a measurement requires more than a number).
• Deploy in constrained environments where fabs allow it using on-prem and air-gapped inference, selecting local or hosted models based on latency, cost, and data-control needs, and supporting release processes that survive customer change control.
• Own rigorous evaluation using golden task sets, tool-call correctness checks, comparisons between agent-built and expert-built recipes against the TMU budget, plus regression tests for every model or prompt update.
• Keep measurement science sound through surrogate models, parameter inference, tool calibration, and calibrated uncertainty, including domain shift across tools, layers, and fabs.
• Set standards and lead engineering execution with design and code reviews, mentorship of mid-level and junior engineers, and quality improvements through unit testing, automated CI/CD testing, refactoring, and static and dynamic analysis.
• Use AI-assisted development responsibly by keeping human review and automated validation in place (static analysis, security scanning, and requirement-based checks), then improving team workflows where gaps are found.
• Translate customer issues into deployable agents and communicate results clearly to engineers who will compare outcomes against reference techniques.
• Set the vision and mentor the AI and ML engineers around you.

What you bring

• MS or PhD in computer science, electrical engineering, physics, applied mathematics, or equivalent hands-on experience.
• 6+ years of applied ML/AI experience in production, including 3+ years in technical leadership (leading projects or owning technical direction for a significant AI system).
• Experience defining AI strategy and system architecture and getting a technical vision adopted across engineering.
• Hands-on depth in LLM-based systems: agents, tool calling, retrieval, and prompt/model evaluation shipped to real users with monitoring after release.
• Strong ML foundations, including deep learning in PyTorch, with experience in regression, anomaly detection, and calibrated uncertainty.
• Working knowledge of MLOps: reproducible training, dataset/model versioning, model registries, CI/CD, monitoring, and GPU-aware inference optimization with agile/DevOps practices.
• Track record of evaluating AI systems rigorously using test sets, metrics, and honest failure analysis.
• Deep Python and strong software engineering fundamentals (data structures, design patterns, and object-oriented design) with a habit of writing clean, tested production code.
• Experience with AI-assisted development tools, including reviewing generated code for errors and incorrect assumptions and turning it into maintainable production systems.
• Understanding of software security, intellectual property, data privacy, and responsible AI in enterprise software.
• Strong analytical and troubleshooting skills for complex systems, with the ability to research and solve hard technical problems independently.
• Experience developing software on Windows with Visual Studio and Azure DevOps.
• Experience owning AI systems in production after launch, including monitoring, diagnosing failures, and shipping updates.
• Clear technical writing and the ability to explain AI decisions to process engineers who need more than “the model decided.”

Technologies you’ll work with

PyTorch, Python, CI/CD, GPU-aware inference optimization, Windows, Visual Studio, Azure DevOps, LLM-based systems, agents, tool calling, retrieval, prompt and model evaluation, MLOps, model registries, monitoring, unit testing, static and dynamic analysis, security scanning, surrogate models, parameter inference, tool calibration, calibrated uncertainty, TMU budget, regression tests, golden task sets.

Benefits

• Health, dental, and vision coverage
• Life and disability insurance
• PTO
• 401(k) with employer match
• Employee Stock Purchase Program (ESPP)
• Wellness initiatives
• Competitive salaries and a comprehensive benefits package

Extra signals we value

• Strong programming skills in C++, C#, or CUDA where the model meets the instrument
• Semiconductor process control experience (metrology, inspection, SPC, yield analytics, or time in a fab)
• Scatterometry or OCD, ellipsometry, reflectometry, interferometry, or e-beam imaging
• Tool-calling and MCP frameworks, multi-agent orchestration, vector and hybrid search, LLM fine-tuning, and on-prem LLM serving
• Security and IP protection for AI systems, including shipping within regulated release processes or with customer data that cannot leave the site
• Physics-informed ML, surrogate modeling, physics-based modeling and numerical simulation, or Bayesian inverse methods (MCMC or variational inference)
• Parallel or high-performance computing
• Experience integrating AI into software development (AI-assisted code review, automated test creation, requirements analysis)

Working here

• This role is on-site by design for close daily collaboration with applications and system engineering.
• Expect regular office time, plus occasional time at a fab or customer site.
• Work is reviewed, documented, and qualified before release, then runs in production lines measuring wafers around the clock.

Compensation & growth

Base salary range: $129,600.00 to $194,400.00 (offered in good faith based on experience, location, and qualifications). Annual bonus opportunities and potential long-term incentives may also be available, tied to company and individual success.

Export compliance note: For certain positions requiring access to technical data, U.S. export licensing review may be necessary for applicants who are not U.S. Citizens, Permanent Residents, or other protected persons under 8 U.S.C. 1324b(a)(3).